Andy Wade wrote:

http://wiki.diyfaq.org.uk/index.php?...Circuit_faults

1. Wire resistance table: the figures seem rather high - they appear to
be circuit resistances (i.e. resistance of the phase-neutral loop) at 70
deg. C, as you would use for calculating voltage drop, or fault current
calculations. In the context of resistance measurements made "cold" you
need the figures at 20 deg. C, which are all in Table 9A of the OSG (p.
158). Also note spelling of "metre" :-)

Well spotted that man!

Indeed I did lift them from the wrong table (6E2 in OSG).

Is it just me, or do you spend ages looking for tables in that book that
you know are there, but can never seem to find when you need them?

You could refine the table by adding two more columns to give
resistances for a single conductor, for the P-N loop (R1+Rn) and for the
P-E loop (R1+R2) of twin & earth cables.

Probably worth doing a simplified version of 9A since the original
resistance table I originally only planed to use on L/N round trip,
since it does add other diagnostic options.

2. Resistance measurements: the method with a 3-pole shorting plug is
invalid unless you disconnect the relevant neutral at the CU. Remember
that N & E are also shorted at the supply end, in the DNO's cut-out for
PME and further back for TN-S, so there are parallel paths which will
invalidate your resistance calculation. I see you do say "one should
always ensure power is turned off at the main switch" but many will
disregard that, taking it only as a safety warning and not realising the
subtle effect it will have on the measurements.

Yup, I had already reworded that a bit, but I agree it is still not
clear about all the implications of not switching off.

3. Disconnected CPC, ring circuit "circuit safety mostly unaffected":
that's a rather sweeping statement and there will be cases where a
dangerous situation exists, due to inadequate thermal protection for a
CPC. The worst case is probably a broken CPC at one end of a long ring,
with the old-type 2.5 T&E which only has a 1.0 mm^2 CPC and with a 30 A
BS 3036 (rewireable) fuse. With an earth fault near the broken far end

Yes good point. I did some checks for modern cables and typical circuit
lengths, and it usually hangs together. However older installs are going
to be more vulnerable.

4. Ring continuity tests: the cross-connection method (per OSG, GN3 and
our FAQ) should be mentioned. It's very convenient, having
cross-connected the ends, to go round and measure resistance between L &
N at all sockets. If all is well you should measure the same resistance
value at all points. Ditto for cross-connection of L & E, except that
the resistance will rise slightly toward the point furthest from the
cross-connection, due to the higher resistance of the 1.5 mm^2 CPC.

Yup, personally I favour that method, although for a simple test here it
may get overly complex since we would need a fairly full discussion on
the variations in readings one will get on spurs etc. (There comes a
point where one is just duplicating swathes of the OSG!)


--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/